Conversation

Notices

  1. Is it my imagination or has EVERYBODY's # instance been running like crap lately?

    Monday, 13-Jun-16 01:00:48 UTC from shitposter.club
    1. @moonman Mines been mostly fine except for the occasional hiccup when someone decides to try to get around my images patch.

      Monday, 13-Jun-16 01:01:38 UTC from community.highlandarrow.com
      1. @moonman I sense with later versions we're starting to see requirement creep beyond the smaller virtual boxes that many instances are on, though.

        Monday, 13-Jun-16 01:03:08 UTC from community.highlandarrow.com
        1. @maiyannah I tried to load a notice on yours and I sat there for like most of a minute waiting for it to load. It did eventually though.

          Monday, 13-Jun-16 01:04:00 UTC from shitposter.club
          1. @moonman Weird, it's fine for me.  Might be your connection?

            Monday, 13-Jun-16 01:04:32 UTC from community.highlandarrow.com
            1. @maiyannah I haven't gon directly to your site for a while so it might be a caching thing it's doing.

              Monday, 13-Jun-16 01:05:09 UTC from shitposter.club
              1. @maiyannah after the first notice it all worked fine.

                Monday, 13-Jun-16 01:05:28 UTC from shitposter.club
            2. @moonman I get 1.69 seconds load (harharhar) average when I run it through my small battery of speed test sites.

              Still a little slow for what it should be but hardly intolerable.

              Monday, 13-Jun-16 01:05:54 UTC from community.highlandarrow.com
              1. @maiyannah @moonman I notice it loads fast on Firefox, but Chrome takes a while to load your instance completely.

                Monday, 13-Jun-16 01:07:49 UTC from shitposter.club
                1. @normandy @moonman Chrome doesn't process asynchronous javascript .. asynchronously, so that's to be expected, whereas IE, and Firefox do.

                  Monday, 13-Jun-16 01:08:28 UTC from community.highlandarrow.com
                  1. @maiyannah @moonman Firefox's javascript engine has been edging out Chrome's for a while now.

                    Monday, 13-Jun-16 01:09:38 UTC from shitposter.club
                    1. @normandy @moonman I wonder if I can work around that in Chrome a bit.  I kind of managed to by accident on the main site and I have no idea how I managed to since it was kind of on a fluke.

                      Monday, 13-Jun-16 01:10:31 UTC from community.highlandarrow.com
                  2. @maiyannah @normandy none of them do unless you're using web workers or service workers. Unless I misunderstand what you're saying. All JS engines are singlethreaded. and can only interrupt a function that has a yield in it.

                    Monday, 13-Jun-16 01:11:58 UTC from shitposter.club
                    1. @maiyannah @normandy oh did you mean ajax streams? I'm a bit out of it.

                      Monday, 13-Jun-16 01:13:42 UTC from shitposter.club
                    2. @moonman @normandy I don't mean executing the javascript, I mean loading them.  When you flag a javascript to load asynchronously, the browser loads all those javascripts at the same time.  If you do not, it loads them sequentially, which increases the load time for the site, obviously (as well as the connection overhead if you aren't using/abusing KEEPALIVE)

                      Monday, 13-Jun-16 01:13:43 UTC from community.highlandarrow.com
                      1. @maiyannah @normandy oh gotcha the async attribute. yeah gotta be careful using that anyhoos since you can't guarantee what shows up first. I think requirejs can handle that, I haven't looked to0 carefully because I write my own naive implemenations based on globals.

                        Monday, 13-Jun-16 01:15:33 UTC from shitposter.club
              2. @moonman ./api/statusnet/conversation/72838.json?count=100&_=1465779905037 is the longest load point by far though so the API may be bottlenecking/slowing it down some.

                Monday, 13-Jun-16 01:07:44 UTC from community.highlandarrow.com
                1. @moonman Of that 1.69s it's waiting 1.56s of that for the API so .. yeah, it probably is.

                  Monday, 13-Jun-16 01:09:37 UTC from community.highlandarrow.com
    2. @moonman I haven't noticed any particular problems on mine. I did hear that some instances were moved around today, though.

      Monday, 13-Jun-16 01:03:46 UTC from gs.kawa-kun.com at 47°40'7"N 122°22'40"W
    3. Are you sure you're not accidentally doing everything through #?

      Monday, 13-Jun-16 06:50:25 UTC from social.umeahackerspace.se at 63°49'42"N 20°15'34"E